Microsoft to Unveil Its First AI‑Focused Chip ‘Athena’ at Ignite Conference

On October 7, reports citing informed sources said Microsoft plans to launch its first chip designed specifically to support artificial intelligence (AI) at next month's annual developer conference. This effort, years in the making, could help Microsoft reduce its reliance on Nvidia AI chips, which have been in short supply due to surging demand.

Microsoft's chip, similar to Nvidia's graphics processing units (GPUs), is built for data‑center servers that train and run large language models—the software behind conversational AI such as OpenAI’s ChatGPT. Microsoft’s data‑center servers currently use Nvidia GPUs to support AI features for cloud customers, including OpenAI and Intuit, as well as Microsoft productivity applications.

The chip, codenamed “Athena,” may be unveiled on November 14 at Microsoft’s Ignite conference in Seattle. Athena is expected to compete with Nvidia’s flagship H100 GPU for AI acceleration in data centers. The custom chip has already been secretly tested by Microsoft and its partner OpenAI.

Microsoft began developing Athena around 2019, aiming to cut costs and gain bargaining power with Nvidia. Azure now relies on Nvidia GPUs for AI workloads used by Microsoft, OpenAI, and other cloud customers. With Athena, Microsoft can follow rivals AWS and Google in offering a home‑grown AI chip to cloud users.

Performance details of Athena are still unclear, but Microsoft hopes the chip can match Nvidia’s H100. Although many companies tout superior hardware and cost efficiency, Nvidia GPUs remain the preferred choice for AI developers thanks to the CUDA platform. Attracting users to new hardware and software will be critical for Microsoft.

A self‑developed AI chip could also reduce Microsoft’s dependence on Nvidia GPUs amid tight supply. Reports say that after deepening its partnership with OpenAI, Microsoft ordered at least several hundred thousand Nvidia chips to support OpenAI’s products and research. Using its own chip could save substantial costs.

OpenAI may also be considering reducing its reliance on Microsoft and Nvidia chips. Recent reports indicate the AI research lab is exploring building its own AI chip, and its hiring notices suggest it is looking for talent to evaluate and co‑design AI hardware.

While Microsoft and other cloud providers do not plan to stop buying Nvidia GPUs immediately, persuading their cloud customers to adopt internal chips over Nvidia GPUs could be economically beneficial in the long run. Microsoft is also working closely with AMD on the upcoming AI chip MI300X. As AI workloads surge, this diversified approach offers multiple options. Competitors in cloud computing are also adopting similar strategies to avoid vendor lock‑in.

Amazon and Google have strategically integrated their AI chip strategies into their cloud offerings. Amazon funds Anthropic, a competitor to OpenAI, on the condition that Anthropic uses Amazon’s AI chips, Trainium and Inferentia. Meanwhile, Google Cloud announced that AI image developers Midjourney and Character AI are using the company’s Tensor Processing Units.

As AI chips become a critical component of data centers, betting on this field could yield high returns. With Athena, Microsoft joins the race to capture market share in the rapidly evolving AI‑chip arena, offering cloud customers more choices and charting a more independent path for next‑generation AI infrastructure.